压缩感知算法在阵列测向中的相干信号分辨

现有算法在进行信号重建时,需要原始信号的稀疏信息来控制初始原子集的大小和算法迭代的最大次数,这削弱了重建精度,增加了计算复杂度,限制了其实际应用能力。为了克服这个问题,提出了基于正交匹配追踪算法的改进算法。该算法将阵元个数作为假设稀疏度,在稀疏度未知的初始阶段扩展初始原子集。最后通过对测量信号的幅度进行阈值过滤以实现信号的分辨和信源数的估计。为解决测向人员面对监测软件所指示的多个信号示向度中,无法分辨哪些指示方向是同一个信号源所产生的折反射信号的问题,首次提出了基于压缩感知算法的相干关系分析和相干信号分辨的模型。实测数据结果表明,该算法在不同环境下可以清晰分辨相干信号,并且能够适应多种实验平台。     

Application of scanning tunnelling microscopy to the quality control of tungsten carbide (WC-Co) ball bearings

Scanning tunnelling microscopy (STM) has been applied to the topographical study of tungsten carbide (with 6% Co) ball bearings in air. It is shown that, for this material, scanning tunnelling microscopy can provide not only general topographical information on the surface microtexture with a resolution of nearly 0.1 nm if required, but also quantitative information for the measurement of surface roughness. Thus, scanning tunnelling microscopy has utility as a quality control tool of particular relevance in the submicron range. The STM topographical results from specific tungsten carbide ball bearings are presented, and the limitations and applicability of the STM technique when used as a quality control tool are discussed.     

基于国产处理器增量式实时交通流预测算法及实现

针对城市交通难以处理大量数据且实时性差等问题,提出了根据增量式城市交通流数据预测拥堵情况的一种基于国产处理器的L-BFGS(limited-memory BFGS)算法。该算法通过存储向量序列计算Hessian矩阵,改进Two-Loop算法求下降方向,在Spark集群中并行处理时收敛速度快,适用于实时性要求强的城市交通场景。实验结果证明,L-BFGS预测算法完全可以在国产平台上对大规模的实时交通数据流进行快速建模、预测,在改善城市交通管理水平提供有效支撑的同时也丰富了国产芯片的应用领域。     

地震波层析成像探测技术与水平定向钻拉管技术在地下障碍物探测中的应用

顶管施工属于非开挖施工方法,是一种不开挖或者少开挖的管道埋设施工技术,该项技术彻底解决了管道埋设施工中对城市建筑物的破坏和道路交通的堵塞等难题,在稳定土层和环境保护方面凸显其优势。但是,在顶管施工过程中一旦遇到地下障碍物,将导致顶管顶力突然增大,顶管顶进方向偏离设计轴线,严重影响顶管施工的顺利进行。福州市某污水顶管施工需要穿越河道,通过前期资料收集发现,河道两侧可能存在护岸抛石等地下障碍物。为了查明护岸抛石的分布情况,综合应用了地震波层析成像探测技术与水平定向钻拉管技术进行探测,既为后期顶管施工的顺利进行提供了参考依据,又降低了探测成本,提高了探测效率。     

Efficient coordinated control of regenerative braking with pneumatic anti-lock braking for hybrid electric vehicle

Urban bus has to start and stop frequently due to typical urban traffic conditions, which, however, can be put to good use by regenerative braking. Regenerative braking is a key technology which not only improves vehicle’s fuel economy in mild braking, but also ensures vehicle safety in emergency braking conditions. Because of the inherent limitations of traditional braking system in recycling energy, it is necessary to change its structure to decouple the brake pressure and the brake pedal force. To solve this problem, a compromise design combining traditional pneumatic braking system with brake-by-wire (BBW) system is adopted in this paper on parallel hybrid electric bus. With the transformed braking system, an efficient coordinated control strategy is proposed to solve the problem caused by the different response speeds of pneumatic braking and regenerative braking. The proposed control strategy is carried out, where the road condition varies and different control methods are adopted. Results show that the adopted braking system and the proposed coordinated control strategy are suitable for different roads, and effective in recovering energy and ensuring vehicle safety. At the same time, shorter braking distance and better control of slip ratio verify the performance of MPC compared with a logical threshold-based control. Therefore, this study may offer a useful theoretical reference to the choice of braking system and braking control strategy design in hybrid electric vehicle (HEV).     

Optimal Design of Two-Chamber Gas Distributor with CFD Approach

Computational fluid dynamics( CFD) is used to investigate a new type of two-chamber natural gas distributor,which has a natural gas inlet and nine nozzle outlets. The uniformity at the outlet of distributor is practice proven to have significant degree influence on its comprehensive performance. To improve the uniformity at the nozzles of the gas distributor,CFD modeling with the RNG k-ε turbulence model is undertaken to understand the mass flow rate of nozzles with reference to different length of chambers and the most optimal length is obtained. The internal flow pattern of the natural gas distributor is analyzed. It is found that the local maximum deviation of the nozzle outflow rate increases with the increase of chambers length when the length is more than 64 mm. The results provide useful suggestions for the optimal design of two-chamber natural gas distributor.     

Zn对AM60B镁合金显微组织和力学性能的影响

以AM60B为基体合金,采用ICP、OM、SEM、EDS和拉伸试验机等研究不同Zn含量对AM60B合金组织和力学性能影响.研究结果表明:随着Zn元素含量的增加,AM60B镁合金的组织主要由α-Mg基体相,Mg2Zn和Al9Mn11相组成,其抗拉强度、屈服强度和硬度先增加后减小,并在Zn元素含量达到4%时达到峰值,分别为248 MPa、123 MPa和70.4HV.其拉伸断口为准解理断口,该合金的强化机制为细晶强化、固溶强化和第二相析出强化,其中细晶强化起主导作用.     

构建基于能力的新员工培训项目开发体系——发散思维、创新模式、注重实效、铸就品牌

为适应国家电网公司当前改革和发展的新形势,加快人才队伍建设,优化新入职员工培训模式,强化培训项目开发工作,构建了基于能力的新员工培训项目开发体系,将调研、筹划、执行、评估等一系列工作纳入该闭环体系中,提升了工作效率,实现了培训质量的持续提升。     

顶升调正纠偏法在储油罐安装过程中的应用

介绍了立式钢制储油罐倒装法施工过程中,采用整体液压顶升油罐后注入沥青砂的调正纠偏具体做法,解决了因底板边缘板局部在罐自身重力作用下出现凸凹变形,导致底层壁板与罐底板角焊缝无法组对焊接的问题,对解决大型储油罐安装过程中的此类问题有一定的指导意义.     

结合PCA的SVM方法在学校财务风险预警中的运用

利用支持向量机方法对中职学校财务风险进行预测,设置22个评价指标,运用主成分分析降维处理数据,然后通过Libsvm进行实验。通过对评价指标的主成分分析,发现前8个主分量的累计贡献率为89.18%,占有绝对重要的比重。选取前8个主分量为支持向量机的输入,最后得到5组实验数据,平均预测分类精度为93.75%,表明支持向量机方法在财务风险预测中的可行性和实用性。