融合 LSTM 和 PPO 算法的移动机器人视觉导航

为提高移动机器人在无地图情况下的视觉导航能力,提升导航成功率,提出了一种融合长短期记忆神经网络( long short term memory, LSTM)和近端策略优化算法(proximal policy optimization, PPO)算法的移动机器人视觉导航模型。 首先,该模型融 合 LSTM 和 PPO 算法作为视觉导航的网络模型;其次,通过移动机器人动作,与目标距离,运动时间等因素设计奖励函数,用以 训练目标;最后,以移动机器人第一视角获得的 RGB-D 图像及目标点的极性坐标为输入,以移动机器人的连续动作值为输出, 实现无地图的端到端视觉导航任务,并根据推理到达未接受过训练的新目标。 对比前序算法,该模型在模拟环境中收敛速度更 快,旧目标的导航成功率平均提高 17. 7%,新目标的导航成功率提高 23. 3%,具有较好的导航性能。     

Intrusion detection and prevention system for an IoT environment

Internet of Things (IoT) security is the act of securing IoT devices and networks. IoT devices, including industrial machines, smart energy grids, and building automation, are extremely vulnerable. With the goal of shielding network systems from illegal access in cloud servers and IoT systems, Intrusion Detection Systems (IDSs) and Network-based Intrusion Prevention Systems (NBIPSs) are proposed in this study. An intrusion prevention system is proposed to realize NBIPS to safeguard top to bottom engineering. The proposed NBIPS inspects network activity streams to identify and counteract misuse instances. The NBIPS is usually located specifically behind a firewall, and it provides a reciprocal layer of investigation that adversely chooses unsafe substances. Network-based IPS sensors can be installed either in an inline or a passive model. An inline sensor is installed to monitor the traffic passing through it. The sensors are installed to stop attacks by blocking the traffic using an IoT signature-based protocol.     

极地环境含风氢储混合微电网容量优化配置

极地环境复杂多样,常年低温且伴有极夜现象。为了合理配置极地环境微电网系统容量,提高极地环境微电网系统的供电可靠性,文中以中国南极科考站——中山站为研究对象,根据极地环境特性,提出利用富余风电制氢,并配置蓄电池储能装置的风氢储微电网容量优化配置模型。综合考虑极地低温对风机出力、蓄电池容量以及电力负荷的实时影响,以年度平均成本最小为目标,全年负载缺电率为约束,对极地环境下微电网供电系统的容量进行优化配置。算例仿真结果表明,风氢储系统不仅可以提高系统经济性和供电可靠性,还能有效减少能源浪费。     

Stock-Price Forecasting Based on XGBoost and LSTM

Using time-series data analysis for stock-price forecasting (SPF) is complex and challenging because many factors can influence stock prices (e.g., inflation, seasonality, economic policy, societal behaviors). Such factors can be analyzed over time for SPF. Machine learning and deep learning have been shown to obtain better forecasts of stock prices than traditional approaches. This study, therefore, proposed a method to enhance the performance of an SPF system based on advanced machine learning and deep learning approaches. First, we applied extreme gradient boosting as a feature-selection technique to extract important features from high-dimensional time-series data and remove redundant features. Then, we fed selected features into a deep long short-term memory (LSTM) network to forecast stock prices. The deep LSTM network was used to reflect the temporal nature of the input time series and fully exploit future contextual information. The complex structure enables this network to capture more stochasticity within the stock price. The method does not change when applied to stock data or Forex data. Experimental results based on a Forex dataset covering 2008–2018 showed that our approach outperformed the baseline autoregressive integrated moving average approach with regard to mean absolute error, mean squared error, and root-mean-square error.     

Simulation analysis and experimental verification of coal suction characteristics of the new railway tunnel fallen coal dust collection device北大核心CSCD

Due to air turbulence, large areas of coal will fall when the special coal-transportation trains pass the tunnel exits and entrances. Aiming at the problems of low efficiency and high cost of manual cleaning for long distance coal cleaning in the tunnel, a new railway tunnel fallen coal dust collection device which was composed of a main conveying coal feeding pipe and multiple branch pipes of coal suction was designed. It was used to clean the small particles and lightweight railway tunnel fallen coal. Firstly, the gas-solid two-phase flow model based on the Euler-Lagrange approach for the design of the main conveying coal feeding pipe was established in the coal conveying pipelines. Secondly, the effect of the coal particles' incident angle and multiple branch pipe spacing on the main coal conveying pipe flow field, which was based on Fluent finite element simulation software, was studied. What was more, the optimal angle of incidence and the optimal value of the number of branch coal suction pipe, which was installed on the main conveying pipe, were analyzed. Finally, the finite element simulation was verified by field test. Simulation and experimental results showed that it was more conducive to the railway tunnel fallen coal transportation when coal particles' incident angle was less than 45° and the branch pipe spacing was in the vicinity of 750 mm. For that when incident angle was less than 45°, the main conveying coal pipe pressure-drop became weaker and particle flow could obtain large horizontal transport velocity. And when the branch pipe spacing was in the vicinity of 750 mm, the horizontal transport velocity had a smaller fluctuation range and the transportation of coal was larger than that of the other groups. The research results are of great significance to improve the structure of the main conveying coal pipe, increase the efficiency of tunnel coal conveying and optimize the railway tunnel coal dust collection device.     

从数字制造到智能制造的关键技术途径研究

在深入研究智能制造的内涵及关键技术的基础上,提出了我国从数字制造到智能制造的三大发展模式,以及实现从数字制造到智能制造发展的具体技术途径。针对典型行业的生产特点,提出了从数字制造到智能制造发展的技术路线图,为推动我国制造业从数字制造到智能制造的发展提供技术途径的指引。     

The influence of particle geometry and the intermediate stress ratio on the shear behavior of granular materials

The behavior of granular materials is very complex in nature and depends on particle shape, stress path, fabric, density, particle size distribution, amongst others. This paper presents a study of the effect of particle geometry (aspect ratio) on the mechanical behaviour of granular materials using the discrete element method (DEM). This study discusses 3D DEM simulations of conventional triaxial and true triaxial tests. The numerical experiments employ samples with different particle aspect ratios and a unique particle size distribution (PSD). Test results show that both particle aspect ratio (AR) and intermediate stress ratio \((b=({\upsigma }_{2}'-{\upsigma }_{3}')/({\upsigma }_{1}'-{\upsigma }_{3}'))\) affect the macro- and micro-scale responses. At the macro-scale, the shear strength decreases with an increase in both aspect ratio and intermediate stress ratio b values. At the micro-scale level, the fabric evolution is also affected by both AR and b. The results from DEM analyses qualitatively agree with available experimental data. The critical state behaviour and failure states are also discussed. It is observed that the position of the critical state loci in the compression \((e-p')\) space is only slightly affected by aspect ratio (AR) while the critical stress ratio is dependent on both AR and b. It is also demonstrated that the influence of the aspect ratio and the intermediate stress can be captured by micro-scale fabric evolutions that can be well understood within the framework of existing critical state theories. It is also found that for a given stress path, a unique critical state fabric norm is dependent on the particle shape but is independent of critical state void ratio.     

Mechanical characterization and modeling of the deformation and failure of the highly crosslinked RTM6 epoxy resin

The nonlinear deformation and fracture of RTM6 epoxy resin is characterized as a function of strain rate and temperature under various loading conditions involving uniaxial tension, notched tension, uniaxial compression, torsion, and shear. The parameters of the hardening law depend on the strain-rate and temperature. The pressure-dependency and hardening law, as well as four different phenomenological failure criteria, are identified using a subset of the experimental results. Detailed fractography analysis provides insight into the competition between shear yielding and maximum principal stress driven brittle failure. The constitutive model and a stress-triaxiality dependent effective plastic strain based failure criterion are readily introduced in the standard version of Abaqus, without the need for coding user subroutines, and can thus be directly used as an input in multi-scale modeling of fibre-reinforced composite material. The model is successfully validated against data not used for the identification and through the full simulation of the crack propagation process in the V-notched beam shear test.     

Redox and pH Dual Responsive Polymer Based Nanoparticles for In Vivo Drug Delivery

Responsive nanomaterials have emerged as promising candidates as drug delivery vehicles in order to address biomedical diseases such as cancer. In this work, polymer‐based responsive nanoparticles prepared by a supramolecular approach are loaded with doxorubicin (DOX) for the cancer therapy. The nanoparticles contain disulfide bonds within the polymer network, allowing the release of the DOX payload in a reducing environment within the endoplasm of cancer cells. In addition, the loaded drug can also be released under acidic environment. In vitro anticancer studies using redox and pH dual responsive nanoparticles show excellent performance in inducing cell death and apoptosis. Zebrafish larvae treated with DOX‐loaded nanoparticles exhibit an improved viability as compared with the cases treated with free DOX by the end of a 3 d treatment. Confocal imaging is utilized to provide the daily assessment of tumor size on zebrafish larva models treated with DOX‐loaded nanoparticles, presenting sustainable reduction of tumor. This work demonstrates the development of functional nanoparticles with dual responsive properties for both in vitro and in vivo drug delivery in the cancer therapy.     

Surface Energy and Surface Stability of Ag Nanocrystals at Elevated Temperatures and Their Dominance in Sublimation‐Induced Shape Evolution

The surface energy and surface stability of Ag nanocrystals (NCs) are under debate because the measurable values of the surface energy are very inconsistent, and the indices of the observed thermally stable surfaces are apparently in conflict. To clarify this issue, a transmission electron microscope is used to investigate these problems in situ with elaborately designed carbon‐shell‐capsulated Ag NCs. It is demonstrated that the {111} surfaces are still thermally stable at elevated temperatures, and the victory of the formation of {110} surfaces over {111} surfaces on the Ag NCs during sublimation is due to the special crystal geometry. It is found that the Ag NCs behave as quasiliquids during sublimation, and the cubic NCs represent a featured shape evolution, which is codetermined by both the wetting equilibrium at the Ag–C interface and the relaxation of the system surface energy. Small Ag NCs (≈10 nm) no longer maintain the wetting equilibrium observed in larger Ag NCs, and the crystal orientations of ultrafine Ag NCs (≈6 nm) can rotate to achieve further shape relaxation. Using sublimation kinetics, the mean surface energy of Ag NCs at 1073 K is calculated to be 1.1–1.3 J m^?2.