Progress and development of particle jet drilling speed-increasing technology and rock-breaking mechanism for deep well

Increasing drilling speed and efficiency of hard formation for deep and ultra-deep well is one of the international recognized drilling problems and key technologies to be tackled urgently. Particle jet impact drilling technology is an efficient non-contact rock-breaking method to overcome slow drilling speed, which has great development and application potential in drilling speed-increase of hard formation and deep well. High efficiency drilling technology and rock-breaking speed-increase mechanism in high temperature, high pressure and high hardness formations of deep and ultra-deep wells were mainly focused and keynoted in this paper. With extensive investigation of domestic and foreign literature, the working principle, key technical devices, deep-well-rock mechanical characteristic, unconventional constitutive model and rock-breaking mechanism of particle jet impact drilling technology were analyzed, which proved the feasibility and high efficiency for deep and hard stratum, and also, dynamic failure mechanism of rock needs to be elaborated by constructing the constitutive model with high temperature and pressure. Meanwhile, the major problems to be solved at present and development direction future were summarized, which mainly included: miniaturization of drilling equipment and individualization of drilling bit; optimization of jet parameters and the evaluation method of rock-breaking effect; establishment of mechanical property and unconventional constitutive model of deep-well-rock; rock-breaking mechanism and dynamic response under particle jet coupling impact. The research can help for better understanding of deep-well drilling speed-increasing technology and also promote the development and engineering application of particle jet impact drilling speed-increase theory and equipment.

Implementation of a DPU-Based Intelligent Thermal Imaging Hardware Accelerator on FPGA

Thermal imaging has many applications that all leverage from the heat map that can be constructed using this type of imaging. It can be used in Internet of Things (IoT) applications to detect the features of surroundings. In such a case, Deep Neural Networks (DNNs) can be used to carry out many visual analysis tasks which can provide the system with the capacity to make decisions. However, due to their huge computational cost, such networks are recommended to exploit custom hardware platforms to accelerate their inference as well as reduce the overall energy consumption of the system. In this work, an energy adaptive system is proposed, which can intelligently configure itself based on the battery energy level. Besides achieving a maximum speed increase that equals 6.38X, the proposed system achieves significant energy that is reduced by 97.81% compared to a conventional general-purpose CPU.

Evaluation of Rutting in Conventional and Rubberized Asphalt Mixes Using Numerical Modeling Under Repeated Loads

This research aimed to predict the permanent deformation (rutting) in conventional and rubberized asphalt mixes under repeated load conditions using the Finite Element Method (FEM). A three-dimensional (3D) model was developed to simulate the Wheel Track Testing (WTT) loading. The study was conducted using the Abaqus/Standard finite element software. The pavement slab was simulated using a nonlinear creep (time-hardening) model at 40°C. The responses of the viscoplastic model under the influence of the trapezoidal amplitude of moving wheel loadings were determined for different speeds and numbers of cycles. The results indicated that a wheel speed increase from 0.5Km/h to 1.0Km/h decreased the rut depth by about 22% and 24% in conventional and rubberized asphalt mixes, respectively. Moreover, increasing the number of cycles from 7,500 (15,000 passes) to 15,000 (30,000 passes) under constant speed increased the rut depth by about 25% and 30% in conventional and rubberized asphalt mixes, respectively. Furthermore, the addition of Crumb Rubber (CR) to the asphalt reduced its rut depth by 55% compared to conventional asphalt.

Response analysis of orthotropic steel deck pavement based on interlayer contact bonding condition

In this research the interlayer contact condition was considered between the adjacent layers of orthotropic steel deck pavement, and an interface contact bonding model was applied to simulate the interlayer bonding condition and evaluate the response of deck pavement under vehicle loads. An advantage of this model is that it can simulate not only the full-bond condition but also the debonding condition at somewhere between adjacent layers. The responses of the orthotropic steel deck pavement were calculated and analyzed by the model, and it found that this model is reasonable and credible to evaluate the responses of the deck pavement comparing with the previous researches. The full-bond condition was an ideal condition between adjacent layers, which was prone to underestimate the responses and deformation of the deck pavement. Moreover, the position and size of the disengaging area have a notable influence on the tensile strain at the top of SMA layer and the bottom of GA layer, and the tensile strain of them also increase with the increase of the disengaging area. Finally, the responses of the steel deck pavement changed obviously when the vehicle speed increase, so the suitable speed limit may reduce the responses and deformation for prolonging the service life of the orthotropic steel deck pavement.

Research on Accessibility and Equity of Urban Transport Based on Multisource Big Data

Daily travel is an important means for everyone to obtain the right to development. With the development of the economy and the progress of the times, the equalization of public infrastructure has become an important concern. The accessibility and fairness of transportation have become a hot topic of research in various fields. To promote transport equity and formulate more reasonable transport planning and policies, this paper takes the Kunshan city as the research object, based on the mobile phone signaling data and the travel time consumption data from the application programming interface (API) of Gaode Map, using weighted average accessibility and the Theil index to investigate the accessibility and equity of public transport and car traffic in the Kunshan city. The study found that the accessibility of public transport is lower than that of car transport in the same research unit, but the equity of public transport is better than that of car transport, that is, the public transport is fair and the efficiency is neglected. In the same mode of transportation, equity presents a high four-week low distribution in the central urban area, and the spatial equity difference is mainly caused by the difference in accessibility levels between cell units. According to the research conclusions, it is recommended that Kunshan further optimize the spatial layout of public transportation infrastructure and adopt measures such as bus speed increase to achieve equity and efficiency and improve the competitiveness of public transportation.

Classification of helicopter’s typical flight state based on threshold

Dividing the 37 flying state of a certain line number helicopter. Firstly, dividing the helicopter rotation and single-engine flight. Secondly, performing preliminary state division for the remaining samlpes, the specific division of yaw angle, helicopter flight altitude and indicated air speed are different states, the least squares polynomial method is used for smoothing respectively. Calculating the extreme value of each parameter data, with the difference value of the extreme value of the parameter data being less than 10 as the limiting condition, dividing the original data segment into non-turning, level flight and steady speed state. The remaining sampling points are in the state of unsteady turning and non-level flight. Taking the difference value 0 as the limiting condition, further divide the non-steady speed and non-level flight state. Dividing the state of turning and non-turning, level flight, ascent and descent, steady speed, increase speed and deceleration state, which is the preliminary division state. Finally, dividing the near-ground and non-near-ground, classifying the helicopter status according to the height threshold, and analyze the accuracy of the classification results. The results show that this method is versatile, can quickly divide helicopters with different flight complexity, and has high accuracy.

Prediction of Gastric Cancer-Related Proteins Based on Graph Fusion Method

Gastric cancer is a common malignant tumor of the digestive system with no specific symptoms. Due to the limited knowledge of pathogenesis, patients are usually diagnosed in advanced stage and do not have effective treatment methods. Proteome has unique tissue and time specificity and can reflect the influence of external factors that has become a potential biomarker for early diagnosis. Therefore, discovering gastric cancer-related proteins could greatly help researchers design drugs and develop an early diagnosis kit. However, identifying gastric cancer-related proteins by biological experiments is time- and money-consuming. With the high speed increase of data, it has become a hot issue to mine the knowledge of proteomics data on a large scale through computational methods. Based on the hypothesis that the stronger the association between the two proteins, the more likely they are to be associated with the same disease, in this paper, we constructed both disease similarity network and protein interaction network. Then, Graph Convolutional Networks (GCN) was applied to extract topological features of these networks. Finally, Xgboost was used to identify the relationship between proteins and gastric cancer. Results of 10-cross validation experiments show high area under the curve (AUC) (0.85) and area under the precision recall (AUPR) curve (0.76) of our method, which proves the effectiveness of our method.

Quick and robust feature selection: the strength of energy-efficient sparse training for autoencoders

Major complications arise from the recent increase in the amount of high-dimensional data, including high computational costs and memory requirements. Feature selection, which identifies the most relevant and informative attributes of a dataset, has been introduced as a solution to this problem. Most of the existing feature selection methods are computationally inefficient; inefficient algorithms lead to high energy consumption, which is not desirable for devices with limited computational and energy resources. In this paper, a novel and flexible method for unsupervised feature selection is proposed. This method, named QuickSelection (The code is available at: https://github.com/zahraatashgahi/QuickSelection), introduces the strength of the neuron in sparse neural networks as a criterion to measure the feature importance. This criterion, blended with sparsely connected denoising autoencoders trained with the sparse evolutionary training procedure, derives the importance of all input features simultaneously. We implement QuickSelection in a purely sparse manner as opposed to the typical approach of using a binary mask over connections to simulate sparsity. It results in a considerable speed increase and memory reduction. When tested on several benchmark datasets, including five low-dimensional and three high-dimensional datasets, the proposed method is able to achieve the best trade-off of classification and clustering accuracy, running time, and maximum memory usage, among widely used approaches for feature selection. Besides, our proposed method requires the least amount of energy among the state-of-the-art autoencoder-based feature selection methods.

Response Surface Mesh with the Outer Input Method

Artificial intelligence in general and optimization tasks applied to the design of aerospace, space,and automotive structures, rely on response surfaces to forecast the output of functions, and are vital part of these methodologies. Yet they have important limitations, since greater precisions require greater data sets, thus, training or updating larger response surfaces become computationally expensive, sometimes unfeasible. This has been a bottle neck limitation to achieve more promising results, rendering many AI related task with a low efficiency.To solve this challenge, a new methodology created to segment response surfaces is hereby presented. Differently than other similar methodologies, the novel algorithm here presented named outer input method, has a very simple and robust operation. With only one operational parameter, maximum element size, it efficiently generates a near isopopulated mesh for any data set with any type of distribution, such as random, Cartesian, or clustered, for domains with any number of coordinates.Thus, it is possible to simplify the response surfaces by generating an ensemble of response surfaces, here denominated response surface mesh. This study demonstrates how a metamodel denominated Kriging, trained with a large data set, can be simplified with a response surface mesh, significantly reducing its often expensive computation costs> experiments here presented achieved an speed increase up to 180 times, while using a dual core parallel processingcomputer. This methodology can be applied to any metamodel, and metamodel elements can be easily parallelized and updated individually. Thus, its already faster training operation has its speed increased.

Response Surface Mesh with the Outer Input Method

Artificial intelligence in general and optimization tasks applied to the design of aerospace, space,and automotive structures, rely on response surfaces to forecast the output of functions, and are vital part of these methodologies. Yet they have important limitations, since greater precisions require greater data sets, thus, training or updating larger response surfaces become computationally expensive, sometimes unfeasible. This has been a bottle neck limitation to achieve more promising results, rendering many AI related task with a low efficiency.To solve this challenge, a new methodology created to segment response surfaces is hereby presented. Differently than other similar methodologies, the novel algorithm here presented named outer input method, has a very simple and robust operation. With only one operational parameter, maximum element size, it efficiently generates a near isopopulated mesh for any data set with any type of distribution, such as random, Cartesian, or clustered, for domains with any number of coordinates.Thus, it is possible to simplify the response surfaces by generating an ensemble of response surfaces, here denominated response surface mesh. This study demonstrates how a metamodel denominated Kriging, trained with a large data set, can be simplified with a response surface mesh, significantly reducing its often expensive computation costs> experiments here presented achieved an speed increase up to 180 times, while using a dual core parallel processingcomputer. This methodology can be applied to any metamodel, and metamodel elements can be easily parallelized and updated individually. Thus, its already faster training operation has its speed increased.