Shenyang Institute of Automation,Chinese Academy Of Sciences
Not a member yet
23582 research outputs found
Sort by
Cloud Computing Based Demand Response Management using Deep Reinforcement Learning
No full textDemand response is an effective way for ensuring safety and stabilization of power grid by maintaining the balance between the supply and the demand of power grid, and this paper focuses on using electric water heaters for demand response. In addition to considering comfort and price factors as did in previous works, this paper considers the overshoot temperature and its influence on demand response. First, a theoretical model of the heating and cooling processes of the electric water heater is established; second, the demand response process using electric water heaters is analyzed, including the influences of the physical parameters and the settings of electric water heaters on the demand response process; third, a model is established considering the demand response requirement, the comfort of owners of electric water heaters, and the electricity price, simultaneously; fourth, an optimization method based on deep reinforcement learning is proposed for demand response using electric water heaters. Meanwhile, the influence of parameters on the results of demand response is discussed in details. Experimental results show the effectiveness of the proposed method</p
Characterization of microstructural anisotropy using the mode-converted ultrasonic scattering in titanium alloy
No full textThe mode-converted (Longitudinal to Transverse, L-T) ultrasonic scattering was utilized to characterize the microstructural anisotropy on three surfaces of samples cut from the low-scattering and high-scattering regions of a raw titanium alloy Ti-6Al-4V billet, respectively. The L-T ultrasonic measurements were performed in two perpendicular directions using two focused transducers with a 15 MHz center frequency in a pitch-catch configuration. The root mean square (RMS) of ultrasonic scattering was calculated for each L-T measurement and a Gaussian function was used to fit each RMS to determine the RMS amplitude. The ratio of RMS amplitudes for L-T measurements performed in two perpendicular directions was calculated to characterize the microstructural anisotropy on the measured surface of a sample. The results show that the amplitude of L-T ultrasonic scattering is highly dependent on the microstructural anisotropy. The microstructural isotropy was considered on the x-y planes of all samples, while the high anisotropy was seen on the x-z and y-z planes of all low-scattering and high-scattering samples. In addition, the microstructural anisotropy measured on the x-z planes of the low-scattering and high-scattering samples gradually increases and decreases, respectively, from the outside diameter (OD) to the centerline (CL) of the billet. The anisotropy measured on the y-z planes of the low-scattering samples slightly decreases and then increases towards the center, while the anisotropy measured on the y-z planes of the high-scattering samples continuously increases towards the center. The variation of microstructural anisotropy in the titanium alloy Ti-6Al-4V billet with duplex microstructure was quantified with the L-T ultrasonic method and the results agree well with micrographs shown in Ref. [18]. The mode-converted ultrasonic scattering method provides a NDE method to characterize microstructural anisotropy, which can be used as an NDE tool for quality control.</p
Robust semantic SLAM with variable structure
No full text基于深度学习的飞速发展，语义信息逐渐成为SLAM（Simultaneous Location and Mapping）领域的研究热点。由于环境以及传感器本身带来的噪声问题，现有大多数语义SLAM算法所构建的语义地图中存在一些异常点，导致构建的语义地图缺乏一致性，并且影响算法精度。损失函数可以调整对异常点分配的权重，从而抑制异常点的存在。但是大多数语义SLAM算法使用的损失函数本身模型固定，不能很好地适应周围环境噪声的变化。为了解决此问题，提出了一种变结构的鲁棒语义SLAM算法，称为VS-SLAM。采用高斯混合相关熵权重函数作为损失函数，利用其可以通过调整参数，随周围环境噪声变化来改变其模型结构的特点，最大程度地拟合噪声的分布，更有利于降低算法对异常点的权重分配，提高对异常点的鲁棒性。在公开的KITTI数据集上的实验表明，比现有的先进方法有更高的精度。在建图的时间几乎相等的情况下，平均相对平移误差和旋转误差分别降低了5.36%和8.82%，并且构建的语义地图更加具有一致性。</p
Context-wise attention-guided network for single image deraining
No full textIn this paper, we propose a context-wise attention-guided network for single image deraining. Unlike most existing deraining methods, our network exploits underlying complementary information not only across multiple scales but also between levels. Specifically, our network architecture is designed to transmit the inter-level and inter-scale features. To extract guiding information and improve the discriminating ability of context-wise attention-guided network, we propose a net-context-wise attention module to generate attention maps. Following residual learning, the clean image is created by removing the predicted rain streak layer from the rainy input. Experimental results show our method has better performance on public datasets than some state-of-the-art methods.</p
Modeling and evaluation of dynamic degradation behaviours of carbon fibre-reinforced epoxy composite shells
No full textIn this study, a segmented degradation model was established for the first time to predict and evaluate the dynamic degradation characteristics of carbon fibre-reinforced epoxy composite (CFRC) shells accurately by considering temperature variations during heating, maintenance, and cooling. The model is based on the Reddy's high-order shear deformation theory, complex modulus method, and coefficient fitting approach. First, explicit expressions of material parameters and thermal expansion coefficients (TECs) in different thermal degradation stages were proposed, followed by the derivation of the differential equations to solve the structural dynamic degradation characteristics. Moreover, an identification method for the fitting coefficients of the CFRC material was described, which is based on experimental tests and iterative calculations of elastic moduli, loss factors, and TECs at different degradation time points in different thermal degradation stages. Finally, the natural frequencies, damping ratios, and time-domain responses were measured using a novel testing system for the dynamic degradation of specimens subjected to a pulse excitation load to validate the developed model. Theoretical and experimental results indicate that the natural frequencies decreased as the degradation time increases at the heating-up and temperature maintenance stages, whereas they increased at the cooling stage. However, an uptrend in the damping and response behaviours was observed in the first two stages, whereas a downtrend was recognized at the cooling stage.</p
Design and Dynamic Modeling of a Flexible Catcher for Noncooperative Targets
No full text为更好地实现对动态非合作目标的捕获，设计开发了一种多臂式柔性捕获器。这种捕获器的原理类似海葵等生物捕猎的方式，不依赖单个柔性臂的精准夹持而是靠多根臂所构成的臂群实现聚拢、挤压等动作，以完成对目标物体的捕获.基于能量守恒和动量守恒原理对非合作目标物体与柔性臂的碰撞问题进行分析，给出了发生碰撞后柔性臂与目标物体各自的运动参数.为进一步分析柔性臂的动态变形过程，采用多个线性关节和扭转关节的组合对单根柔性臂进行描述，并基于牛顿法对各离散关节进行受力分析，建立了柔性臂的动力学模型。将柔性臂整个变形过程离散为多个微小时间段运动的集合，通过动力学分析得到当前时刻的动力学参量，经过一个微小时间内的运动后即可得到下一时刻各质点的位置，迭代进行上述步骤便得到了柔性臂的动态变形过程。而后，通过实验确定单臂动力学模型的最优参数，并将参数优化后的动力学模型在不同加载情况下与单臂样机进行对比，验证了动力学模型的准确性。最后，在单臂动力学模型的基础上建立包括多根柔性臂的捕获器三维模型，进行非合作目标捕获的仿真与样机实验。结果表明：所设计的臂群式柔性捕获器能够很好地完成动态非合作目标捕获任务，所建立的捕获器三维仿真模型可以基本反映动态捕获过程。</p
An Expectation Maximization based Adaptive Group Testing Method for Improving Efficiency and Sensitivity of Large-Scale Screening of COVID-19
No full textThe pathogen of the ongoing coronavirus disease 2019 (COVID-19) pandemic is a newly discovered virus called severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). Testing individuals for SARS-CoV-2 plays a critical role in containing COVID-19. For saving medical personnel and consumables, many countries are implementing group testing against SARS-CoV-2. However, existing group testing methods have the following limitations: (1) The group size is determined without theoretical analysis, and hence is usually not optimal. This adversely impacts the screening efficiency. (2) These methods neglect the fact that mixing samples together usually leads to substantial dilution of the SARS-CoV-2 virus, which seriously impacts the sensitivity of tests. In this paper, we aim to screen individuals infected with COVID-19 with as few tests as possible, under the premise that the sensitivity of tests is high enough. We propose an eXpectation Maximization based Adaptive Group Testing (XMAGT) method. The basic idea is to adaptively adjust its testing strategy between a group testing strategy and an individual testing strategy such that the expected number of samples identified by a single test is larger. During the screening process, the XMAGT method can estimate the ratio of positive samples. With this ratio, the XMAGT method can determine a group size under which the group testing strategy can achieve a maximal expected number of negative samples and the sensitivity of tests is higher than a user-specified threshold. Experimental results show that the XMAGT method outperforms existing methods in terms of both efficiency and sensitivity.</p
Process Monitoring for Plasticizing Process of Single-base Gun Propellant Based on Mutual Information and MPCA
No full text针对现阶段单基药塑化过程缺乏可靠的渐变式故障检测和异常工况的报警及评价能力等问题，同时传统数据驱动方法难以处理过程中包含的多种线性与非线性关系混合特征，又无法突出变量间耦合相关性的差异，提出一种基于归一化互信息的多向主成分分析故障检测方法。该方法通过归一化互信息刻画过程中多维变量间的复杂耦合关系，并以此对不同维度变量之间的相关性特征进行权值修正，得到体现每个变量与其他维度变量之间耦合作用关系的数据集，并分别建立与其相应的监测模型。最后再利用贝叶斯推理将不同模型的监测结果融合成一组概率指标实现过程监测。试验结果表明了所提方法的有效性，可实现塑化过程多种故障的快速检测和异常工况的监测预警。</p
3D interest point detection using balance-distortion oriented selection
No full textInterest point detection is a challenging problem in 3D objects. Compared to traditional corner detection based on the curvature, this paper proposes a novel method that quantifies the balance and uniformity of local geometric structures based on the distribution of vertex neighborhoods. We first define the neighborhoods of vertices and structure them within the two-ring, instead of constructing the overall mesh, so as to avoid the interference between the neighborhoods of different vertices. Then we introduce the concept balance-distortion to describe the geometric features of the local structure. The experimental results show that the proposed algorithm is robust against noise and invariant to geometric transformation. In addition, compared with the corner detection, more feature points that do not satisfy the balance and direction uniformity are detected, and the distribution of interest point is more uniform.</p
Design and implementation of agent-based robotic system for agile manufacturing: A case study of ARIAC 2021
No full textWe share our experience and lessons learned from participating in the Agile Robotics for Industrial Automation Competition (ARIAC), 2021. We won 1st place in the finals and benefited from an agent-based architecture, task assignment strategy, and modular software programming design method based on the Unified Modeling Language.</p