正则稀疏化的多因子量化选股策略

针对高维度数据集特征之间的复杂性,而传统的L1惩罚项不满足Oracle性质的无偏性,将逻辑回归弹性网（LR-Elastic Net）中的L1惩罚项替换为SCAD（Smoothly Clipped Absolute Deviation）和MCP（Minimax Concave Penalty）惩罚项,分别构建了LR-SCAD和LR-MCP模型,在保留稀疏性的同时满足了无偏性,并利用ADMM（Alternating Direction Method of Multipliers）算法进行求解。通过模拟实验发现,LR-Elastic Net模型能很好地处理特征存在相关性的小样本数据,而LR-SCAD和LR-MCP模型在特征存在相关性的大样本数据中表现较好;建立LR-Elastic Net、LR-SCAD和LR-MCP策略,并应用于沪深300指数成分股数据。回测结果显示,LR-SCAD和LR-MCP策略在股票相关性很强的数据中比LR-Elastic Net策略表现更好。     

基于tPUF的物联网设备安全接入方案

针对现有的物联网设备安全接入方案不适用于资源受限的物联网设备的问题,提出一种基于tPUF的物联网设备安全接入方案。利用物理不可克隆函数技术（Physical Unclonable Function,PUF）,物联网设备不需要存储任何秘密信息,实现设备与认证端的双向认证以及协商会话秘钥;利用可信网络连接技术（Trusted Network Connect,TNC）,完成认证端对物联网设备的身份认证、平台身份认证、完整性认证。安全性分析表明,方案能够有效抵抗篡改、复制、物理攻击等。实验结果表明,相较于其他方案,该方案明显降低了设备的资源开销。     

客户分类下生鲜配送两级路径问题与算法研究

生活水平的提高使得消费者对生鲜产品的需求不断增长,进而促进了冷链物流行业的快速发展。将客户按重要性分为重要客户和普通客户两类,以总配送成本最小为目标,建立考虑客户分类的两级容量有限车辆路径优化模型。提出两阶段启发式算法求解该模型：第一阶段设计改进的遗传-模拟退火算法增强全局搜索能力,其中采用轮盘赌选择机制结合精英保留策略保留优秀个体,部分匹配交叉算子结合自适应交叉率维持种群多样性,Metropolis准则以一定概率接受较差解;第二阶段使用精确方法求解一级配送路径。基于Perboli的Set2算例集和Hemmelmayr的Set5算例集,共30个基准案例,分别将所提出算法与四种现有算法进行对比分析,验证了改进算法的效果,并测试了算法的收敛性。基于模拟数据进行模型分析,验证了所提出模型和算法的有效性和适用性。     

引入源端信息的机器译文自动评价方法研究

机器译文自动评价是机器翻译中的一个重要任务。针对目前译文自动评价中完全忽略源语言句子信息,仅利用人工参考译文度量翻译质量的不足,该文提出了引入源语言句子信息的机器译文自动评价方法: 从机器译文与其源语言句子组成的二元组中提取描述翻译质量的质量向量,并将其与基于语境词向量的译文自动评价方法利用深度神经网络进行融合。在WMT-19译文自动评价任务数据集上的实验结果表明,该文所提出的方法能有效增强机器译文自动评价与人工评价的相关性。深入的实验分析进一步揭示了源语言句子信息在译文自动评价中发挥着重要作用。     

Loop Closure Detection based on Image Covariance Matrix Matching for Visual SLAM

International Journal of Control, Automation and Systems - Loop closure detection is an indispensable part of visual simultaneous location and mapping (SLAM). Correct detection of loop closure can...     

Expectation Maximization Algorithm for Time-delay Output-error Models Based on Finite Impulse Response Method

International Journal of Control, Automation and Systems - In this paper, an output error (OE) model with random time delay is identified by using the expectation maximization (EM) algorithm. Since...     

Fixed-time Group Consensus of Nonlinear Multi-agent Systems via Pinning Control

This paper deals with the fixed-time group consensus problem for multi-agent systems (MASs) subjected to exogenous disturbances. Firstly, two pinning control algorithms are constructed for MASs, which not only reduce the number of controllers but also achieve expected tracking consensus. Secondly, fixed-time group consensus is ensured by utilizing the algebraic graph theory, Lyapunov stability and fixed-time control technique. Finally, simulations are finally given for demonstrate the availability of the derived results.

     

New Robust Stability Criteria for Lur’e Systems with Time-varying Delays and Sector-bounded Nonlinearities

This paper deals with a robust stability problem for uncertain Lur’e systems with time-varying delays and sector-bounded nonlinearities. An improved delay-dependent robust stability criterion is proposed via a modified Lyapunov-Krasovskii functional (LKF) approach. Firstly, a modified LKF consisting of delay-dependent matrices and double-integral items under two delay subintervals is constructed, thereby making full use of the delay and its derivative information. Secondly, the stability criteria can be expressed as convex linear matrix inequality (LMI) via the properties of quadratic function application. Thirdly, to further reduce the conservatism of stability criteria, the quadratic generalized free-weighting matrix inequality (QGFMI) is used. Finally, some numerical examples, including the Lur’e system and the general linear time-delayed system, are presented to show the improvement of the proposed approach.

     

Flocking of Multi-agent Systems with Unknown Nonlinear Dynamics and Heterogeneous Virtual Leader

This paper investigates the flocking control of multi-agent systems with unknown nonlinear dynamics while the virtual leader information is heterogeneous. The uncertain nonlinearity in the virtual leader information is considered, and the weaker constraint on the velocity information measurements is assumed. In addition, a bounded assumption on the unknown nonlinear dynamics is also considered. It is weaker than the Lipschitz condition adopted in the most flocking control methods. To avoid fragmentation, we construct a new potential function based on the penalty idea when the initial network is disconnected. A dynamical control law including a adjust parameter is designed to achieve the stable flocking. It is proven that the velocities of all agents approach to consensus and no collision happens between the mobile agents. Finally, several simulations verify the effectiveness of the new design, and indicate that the proposed method has high convergence and the broader applicability in practical applications with more stringent restrictions.

     

基于粒子群优化BP神经网络的超声波缺陷位置预测

海洋设备的无损检测对于保障设备安全使用至关重要,由于设备所处的海洋环境特殊,要实现对海洋设备缺陷位置的准确估计难度较大.论文提出了一种基于粒子群优化BP神经网络(PSO-BP)预测设备缺陷位置的方法.通过模拟了海洋检测环境,用钢板作为实验对象,用超声波探伤仪对缺陷钢板进行了数据采集,获取了水下钢板的实时数据.通过粒子群算法对BP网络进行优化后对数据进行分析,对比缺陷出现的实际位置和预测位置,证明了粒子群优化的BP神经网络对于超声波检测的缺陷出现位置具有较好的预测效果.