人机协作中人体运动预测方法研究综述北大核心

当今世界制造业智能化发展趋势下,产品制造过程愈加重视机器人的使用和柔性化生产,轻量灵活的协作机器人与人进行协同作业能有效提高工作效率,实现人与机器的优势互补。但人机协作的开放环境中,动态障碍物如人的运动轨迹具有时变性和不确定性,为了使人与机器人的协作过程更安全、流畅,机器人需要感知人类同伴正在进行的动作,预测人体下一步的运动轨迹,并相应地进行自己的运动规划。具体来说,机器人具备早期预测的能力很重要。为了保证人机共享时空下操作人员的安全,使人与机器更高效地协作,介绍了近年来国内外学者针对人机协作安全策略中人体运动预测方法的研究状况,并对其发展趋势和下一步的研究方向进行探讨。     

Seismic response and stability of underground rock caverns: a case study of Baihetan underground cavern complex

The seismic stability of the underground cavern complex, which houses the Baihetan hydropower plant in Yunnan Province, China, currently the world’s 2nd largest underground rock cavern group, is studied in this article. A preliminary performance-based seismic assessment approach specified for underground rock caverns is firstly proposed. The seismic performance objectives are classified. Reference earthquake motions are specified with the determination of the seismic variables. Detailed dynamic response analyses (Method 2B) are conducted based on the parameters given by the cyclic dynamic loading tests with medium strain rate. The seismic response of acceleration, stress, displacement and failure zones are studied. In addition, the seismic spectrum characteristics are analysed with a newly introduced wavelet packet technique. The mechanism of the support measures for seismic stability is also discussed. The assessment for seismic performance of the underground cavern complex is obtained by integrating these results. Serviceability objective of the underground cavern complex under operating basis earthquake may be satisfied, and the safety objective be feasible but with the presence of the proposed reinforcement system. The seismic isolation design is preferred, yet not necessary. And if any seismic isolation design is to be adopted, the frequency absorption range of the isolation material is expected to be 1–4 Hz.     

科研院所产业化道路下的用电方式及节能探讨

详细介绍了科研院所产业化道路下的两部制用电方式和电费计算方法，并结合某科研院所2013年用电情况进行分析，最后结合工作实践，总结了一些关于节能的建议。     

小麦谷朊粉在面包生产中的应用研究

本文就谷朊粉应用于面包生产中对主要工艺因素的影响进行了探讨，并对添加谷朊粉后面团的流变学特性及面包的品质进行了分析研究。     

Cr4+:YAG激光器的研究与应用

Cr^4＋：YAG晶体具有很稳定的物理和化学性质，做为激光器的增益介质采用多种抽运源进行抽运。Cr^4＋：YAG同态激光器的性能稳定、结构紧凑，能产生超短光脉冲并可以对激光输出波长在1300～1600nm间进行调谐，它是光纤通信与量子通信的理想波段，也是人眼的安全波段。     

动态压力光子晶体光纤传感器的研究

光子晶体光纤传感器可广泛用于各种动态压力测量中.文章设计了一种动态压力光子晶体光纤传感器,采用差分平衡方法分析了这种传感器的压力作用原理,讨论了这种传感器的输出信号检测方案,结果表明,该传感器对外界压力作用的响应具有周期性,响应周期与外界压力和传感器敏感元件长度相关.     

三苯基膦稳定的负载钌纳米粒子上对氯硝基苯的选择加氢反应

制备了γ-Al2O3负载的三苯基膦稳定的钌纳米粒子（PPh₃-Ru/Al₂O₃）催化剂,并进行了XRD 和XPS表征。将其用于对氯硝基苯的选择加氢反应,考察了反应压力、温度和时间对催化性能的影响。结果表明,在60 ℃、氢压2 MPa、反应21 h和对氯硝基苯几乎完全转化的条件下,实现了产品对氯苯胺99.7%的选择性。分析表明,脱氯反应的抑制是取得产品高选择性的主要原因,阐明了该催化剂如何有效抑制脱氯反应的发生。     

直序扩频通信系统的研究与System View仿真

扩频通信系统与常规的通信系统相比,具有很强的抗干扰能力,并具有信息隐蔽、多址保密通信等特点,正从军事应用向民用通信发展。理论分析了直接序列扩频通信系统的原理,并用功能强大的仿真软件System View对直接序列扩频通信系统进行建模,仿真分析了系统的抗干扰特性等特点,对扩频通信系统的实现有着重要意义。     

一种评价水泥浆沉降稳定性的新方法——微压力波动测试水泥浆滤失—沉降耦合作用

目前的水泥浆沉降稳定性测试方法缺乏对沉降过程的描述。为了提高水泥浆沉降稳定性测试的准确性与科学性,设计了一种微压力波动测试仪,对滤失—沉降耦合作用的水泥浆柱分段压力波动进行测试,据此评价浆体的上、下密度差,并对不同黏度、温度、降失水剂、固相颗粒和触变性的水泥浆稳定性进行了分析。研究结果表明:①水泥浆沉降过程可分为高速期、低速期、稳定期和失重期,高速期滤失可减小浆体的上、下密度差;②滤失量越大、黏度越低、同类材料水固比越大,密度差越大、沉降稳定性越差;③温度对沉降稳定性的影响因降失水剂的协同作用而呈现出无规律性,高分子聚合物类降失水剂效果强于成膜类;④提高水泥浆触变性可改善浆体的沉降分层现象,胶凝强度的形成时间越短,效果越好。结论认为,较之于常规方法,该评价水泥浆沉降稳定性的新方法,可以对沉降全过程进行描述,并考虑了滤失对沉降的影响,更加贴合于井下的实际情况。     

Deep global-attention based convolutional network with dense connections for text classification

Text classification is a classic task innatural language process (NLP). Convolutional neural networks (CNNs) have demonstrated its effectiveness in sentence and document modeling. However, most of existing CNN models are applied to the fixed-size convolution filters, thereby unable to adapt different local interdependency. To address this problem, a deep global-attention based convolutional network with dense connections (DGA-CCN) is proposed. In the framework, dense connections are applied to connect each convolution layer to each of the other layers which can accept information from all previous layers and get multiple sizes of local information. Then the local information extracted by the convolution layer is reweighted by deep global-attention to obtain a sequence representation with more valuable information of the whole sequence. A series of experiments are conducted on five text classification benchmarks, and the experimental results show that the proposed model improves upon the state of-the-art baselines on four of five datasets, which can show the effectiveness of our model for text classification.