Multi-Equipment Detection Method for Distribution Lines Based on Improved YOLOx-s

The YOLOx-s network does not sufficiently meet the accuracy demand of equipment detection in the autonomous inspection of distribution lines by Unmanned Aerial Vehicle(UAV)due to the complex background of distribution lines,variable morphology of equipment,and large differences in equipment sizes.Therefore,aiming at the difficult detection of power equipment in UAV inspection images,we propose a multi-equipment detection method for inspection of distribution lines based on the YOLOx-s.Based on the YOLOx-s network,we make the following improvements:1)The Receptive Field Block(RFB)module is added after the shallow feature layer of the backbone network to expand the receptive field of the network.2)The Coordinate Attention(CA)module is added to obtain the spatial direction information of the targets and improve the accuracy of target localization.3)After the first fusion of features in the Path Aggregation Network(PANet),the Adaptively Spatial Feature Fusion(ASFF)module is added to achieve efficient re-fusion of multi-scale deep and shallow feature maps by assigning adaptive weight parameters to features at different scales.4)The loss function Binary Cross Entropy(BCE)Loss in YOLOx-s is replaced by Focal Loss to alleviate the difficulty of network convergence caused by the imbalance between positive and negative samples of small-sized targets.The experiments take a private dataset consisting of four types of power equipment:Transformers,Isolators,Drop Fuses,and Lightning Arrestors.On average,the mean Average Precision(mAP)of the proposed method can reach 93.64%,an increase of 3.27%.The experimental results show that the proposed method can better identify multiple types of power equipment of different scales at the same time,which helps to improve the intelligence of UAV autonomous inspection in distribution lines.

Machine Learning Design of Aluminum-Lithium Alloys with High Strength

Due to the large unexplored compositional space,long development cycle,and high cost of traditional trial-anderror experiments,designing high strength aluminum-lithium alloys is a great challenge.This work establishes a performance-oriented machine learning design strategy for aluminum-lithium alloys to simplify and shorten the development cycle.The calculation results indicate that radial basis function(RBF)neural networks exhibit better predictive ability than back propagation(BP)neural networks.The RBF neural network predicted tensile and yield strengths with determination coefficients of 0.90 and 0.96,root mean square errors of 30.68 and 25.30,and mean absolute errors of 28.15 and 19.08,respectively.In the validation experiment,the comparison between experimental data and predicted data demonstrated the robustness of the two neural network models.The tensile and yield strengths of Al-2Li-1Cu-3Mg-0.2Zr(wt.%)alloy are 17.8 and 3.5 MPa higher than those of the Al-1Li4.5Cu-0.2Zr(wt.%)alloy,which has the best overall performance,respectively.It demonstrates the reliability of the neural network model in designing high strength aluminum-lithium alloys,which provides a way to improve research and development efficiency.

A Haze Feature Extraction and Pollution Level Identification Pre-Warning Algorithm

The prediction of particles less than 2.5 micrometers in diameter(PM2.5)in fog and haze has been paid more and more attention,but the prediction accuracy of the results is not ideal.Haze prediction algorithms based on traditional numerical and statistical prediction have poor effects on nonlinear data prediction of haze.In order to improve the effects of prediction,this paper proposes a haze feature extraction and pollution level identification pre-warning algorithm based on feature selection and integrated learning.Minimum Redundancy Maximum Relevance method is used to extract low-level features of haze,and deep confidence network is utilized to extract high-level features.eXtreme Gradient Boosting algorithm is adopted to fuse low-level and high-level features,as well as predict haze.Establish PM2.5 concentration pollution grade classification index,and grade the forecast data.The expert experience knowledge is utilized to assist the optimization of the pre-warning results.The experiment results show the presented algorithm can get better prediction effects than the results of Support Vector Machine(SVM)and Back Propagation(BP)widely used at present,the accuracy has greatly improved compared with SVM and BP.

Exhaust emission characteristics of diesel and biodiesel

The performances of diesel contrasted with that of biodiesel, such as HC and CO and NOx and fume and particles discharge, are studied on. The contents of HC, CO, NOx and flue gas are measured by Japan 163 Gas Chromatograph, USA 48 CO Analyser, Japan MEXA-7200H and FBY-201 flue gas analyzer, respectively. The experimental results shows HC dis-charge of biodiesel B20 decreases with the increase of rate of revolution;NOx discharge of biodiesel B5 and biodiesel B20 increase with the increase of torque;NOx discharge increases with the increase of adding biodiesel at the same torque;Flue gas discharge of biodiesel B5 and biodiesel B20 firstly increase and then decrease with the increase of torque;Particle discharge decreases with the increase of the amount of biodiesel.

Quantum Walk of Two Quantum Particles on One Dimensional System

We study two particle quantum walks on one dimensional chain. Probability distribution of two particle quantum walks is dependent on the initial state, and symmetric quantum walk or asymmetric quantum walk is analogous to that of one particle quantum walk. The quantum correlation probability is much different from classical coincidence probability. The difference reflects quantum interference between two particles.

The development and benefits of metformin in various diseases

Metformin has been used for the treatment of type II diabetes mellitus for decades due to its safety,low cost,and outstanding hypoglycemic effect clinically.The mechanisms underlying these benefits are complex and still not fully understood.Inhibition of mitochondrial respiratory-chain complex I is the most described downstream mechanism of metformin,leading to reduced ATP production and activation of AMP-activated protein kinase(AMPK).Meanwhile,many novel targets of metformin have been gradually discovered.In recent years,multiple pre-clinical and clinical studies are committed to extend the indications of metformin in addition to diabetes.Herein,we summarized the benefits of metformin in four types of diseases,including metabolic associated diseases,cancer,aging and age-related diseases,neurological disorders.We comprehensively discussed the pharmacokinetic properties and the mechanisms of action,treatment strategies,the clinical application,the potential risk of metformin in various diseases.This review provides a brief summary of the benefits and concerns of metformin,aiming to interest scientists to consider and explore the common and specific mechanisms and guiding for the further research.Although there have been countless studies of metformin,longitudinal research in each field is still much warranted.

社会政策执行中的地方政府间竞赛与评奖:地方激励理论及其在改善环境卫生中的实践

本文主要讨论了地方政府之间的竞赛与评奖在推动当地政府投资和在特定社会政策领域公众参与的作用。同时,也比较了地方政府间竞赛与评奖与其他激励模式例如消费者选择和标杆分析法之间的优势和劣势。本文认为,地方政府间竞赛与评奖非常有利于创造一种地方认同感,从而引导服务提供者和社会公众实现社会政策的目标。本文以西宁地区在组织竞赛以改善环境卫生的经验为例,说明竞赛与评奖非常有利于提高公共意识和公众参与。它可以用来鼓励贫困地区参与发展过程。

基于微结构演化的V5Cr5Ti合金塑性本构模型

为合理描述V5Cr5Ti合金的塑性变形行为,论文建立了基于微结构演化的塑性本构模型.首先,采用小尺寸试样开展了V5Cr5Ti合金单轴拉伸试验,并对其在不同应变程度下的微结构演化特征进行了分析.研究发现,影响V5Cr5Ti合金塑性变形行为的主要因素是位错密度演化以及团簇状和弥散析出相.据此建立了位错密度演化方程、组分相含量体积分量演化方程,并考虑团簇状和弥散状第二相对V5Cr5Ti合金流动应力的影响,进一步建立了包括非热应力、热激活应力和弥散相强化应力的流动应力关系式.最后,根据隐式应力更新算法对新模型进行了有限元实现,并与实验结果进行比较,验证了新模型的合理性和预测精度.

Reservoir formation mechanism analysis and deep high-quality reservoir prediction in Yingcheng Formation in Longfengshan area of Songliao Basin, China

Although commercial gas flow was produced in several wells with recent years'exploration of Longfengshan area in Changling fault sag,the formation mechanism and controlling factors for high-quality reservoirs still remained undefined.Here,the Yingcheng tight gas reservoirs of Longfengshan area are used as an example to characterize high-quality reservoir formation mechanism and distribution rules.Based on the thin section,SEM,X-ray diffraction,computed tomography(CT)scanning,burial history,constant-rate mercury penetration and physical properties testing,formation mechanism and controlling factors for high-quality reservoirs were analyzed.Results show the following characteristics.First,the reservoir is dominated by chlorite and laumontite cements,and compaction is the most important factor to control reservoir physical properties.According to this,the reservoir can be divided into compacted tight sandstones,chlorite-cemented sandstones and laumontite-cemented sandstones.Second,the high-quality reservoirs are formed due to early extensive laumontite precipitation and the later dissolution of laumontite by organic acid.Meanwhile,it is found that the distribution of cementation and dissolution exhibits some regulations in sedimentary facies,and the distribution is mainly effected and controlled by the lake water and charging of fresh water.Besides,the distribution model of various types of sandstones was established.Studies over diagenesis and sedimentary facies reveal that the high-quality laumontite-cemented sandstones exist in the outside subaqueous fan-delta of the deep sag in Longfengshan area.These findings have been validated by recent exploration wells which obtained high industrial gas flow.

A design for preparation of textured piezoelectric ceramics based on the phase-field simulation

To solve the problem of directional arrangement of small template particles,we designed a lamination technique for the preparation of dense ceramics with high texture degree based on the phase-field simulation.Referring to the experimental data,the initial microstructures of the template and matrix layers were constructed.The effect of the average length of template on the coarsening behavior of the template layer was investigated in detail.The results suggested that there was a stable stage in the growth process of template grains,which would be conducive to the densification of textured ceramics.This phenomenon has been confirmed by corresponding experiments.In addition,we demonstrated a critical thickness of matrix layer for the preparation of highly textured ceramics by using the template with various average lengths.The grain size of highly textured ceramics could be controlled by adjusting the template size and thickness of matrix layer.