基于TTr1SVD的张量奇异值分解及其在人脸识别上的应用

张量是一种数据组织形式,它的实质是高维数组。很多数据都可以被组织成张量的形式。可以考虑将人脸图像组织成张量的形式。人脸识别过程中最重要的一个环节是特征提取, 后续的匹配识别过程是建立在它的基础上。TTr1SVD是一种新型的张量分解算法,可以认为该算法是矩阵SVD在张量领域的扩展。实际数据库里的图片的图像模态往往是最大的,结合TTr1SVD算法,得到张量的高阶奇异值分解,改变图片的组织形式,可以加速人脸特征的提取。本文基于TTr1SVD的高阶奇异值分解算法,实现人脸特征的提取和识别,并且保持了较好的准确性。实验结果表明,该算法比传统的使用Tensor Toolbox的高阶奇异值分解算法更加灵活高效。     

Numerical investigation for SPR-based optical fiber sensor

In the present work, a theoretical analysis of a surface plasmon resonance (SPR)-based optical fiber sensor is carried out. For the SPR-based optical fiber sensor with four-layer mode (fiber core/metal/sensing layer/sample), the effect of the thickness of metal layer and sensing layer on the spectrum of the transmitted power and wavelength has been studied. Based on the propagation wavevector till the first order expansion of the dispersion function, a detailed explanation for the relationships between the resonance wavelength with the thickness of the metal layer and sensing layer is presented.     

Multi-objective evolutionary simulated annealing optimisation for mixed-model multi-robotic disassembly line balancing with interval processing time

This paper considers the design and balancing of mixed-model disassembly lines with multi-robotic workstations under uncertainty. Tasks of different models are performed simultaneously by the robots which have different capacities for disassembly. The robots have unidentical task times and energy consumption respectively. Task precedence diagrams are used to model the precedence relations among tasks. Considering uncertainties in disassembly process, the task processing times are assumed to be interval numbers. A mixed-integer mathematical programming model is proposed to minimise the cycle time, peak workstation energy consumption, and total energy consumption. This model has a significant managerial implication in real-life disassembly line systems. Since the studied problem is known as NP-hard, a metaheuristic approach based on an evolutionary simulated annealing algorithm is developed. Computational experiments are conducted and the results demonstrate the proposed algorithm outperforms other multi-objective algorithms on optimisation quality and computational efficiency.     

Spectra of dual-overwritten chirped fiber Bragg gratings

The coupled-mode equations for dual-overwritten chirped fiber Bragg gratings (CFBGs) are obtained and numerically calculated. By simulating the reflection spectra of dual-overwritten CFBGs, we have investigated the effects of chirp coefficient, period difference, grating length and index modulating coefficient, on the reflection spectrum and its intensity and full-width-at-half-maximum (FWHM).     

基于人体热释电特征多策略融合的识别方法

针对现实复杂环境中,利用单个或多个热释电传感器对人体目标识别时所表现出的不稳定性和抗干扰性较差问题,提出了一种基于人体热释电特征的多策略融合识别方法。该方法以不确定性理论为基础,首先利用不同空域的传感器,获取人体不同部位的置信度,其次结合不同的融合策略实现对人体目标的识别。融合策略分为两种,第一种策略首先使用同空域融合,然后使用同时域融合,这种策略充分利用了分布式传感网络的特性,融合了最优置信度以实现对人体目标的高精度识别,第二种策略则首先使用同时域融合,然后使用同空域融合。该策略能实现对人体目标的实时识别,但不能保证较高的识别精度。实验结果表明多策略融合既能克服单传感器或多传感器在对人体目标进行识别时,所表现出的不稳定性和不可靠性,又能增强系统识别时的鲁棒性和有效性。与其他融合方法相比突出体现了本文方法的优势。     

分布式光纤光栅测温技术在线监测电缆温度

简要介绍了分布式光纤光栅测温技术的基本原理和特点,探讨该技术在电缆温度在线监测中的应用前景。对光纤光栅温度传感器的封装进行了优化设计,并将其安装在220 kV高压电缆上进行为期一年的现场试验,考察了系统的测温精度、响应时间及运行可靠性等指标,试验结果表明该技术测温误差小、响应时间短、运行可靠且能实现长距离测量,可有效应用于电缆温度在线监测,为电缆载流量的确定提供有效的参考依据,避免电缆火灾事故。最后在工频击穿试验中验证了该技术为电缆绝缘在线监测提供了一种新的思路和方法,值得推广。     

Optimization of a Low-Voltage Load Switch for a Smart Meter Based on a Double Response Surface Model

The existing electric energy meter load switches have issues such as a low electrical life, poor resistance to short-circuit currents, high contact point resistivity and severe heating. Research has thus been conducted on the optimization of the electric energy meter built-in load switch mechanism to establish a double response surface model for the electromagnetic attractive force and reed reaction force. The model introduces a niche fitness sorting strategy and a Gaussian mutation mechanism that are based on the standard particle swarm algorithm and form an improved multi-objective particle swarm optimization algorithm in which the static attractive force is regarded as the optimization objective. The multi-objective optimization of the structural parameters of the electromagnetic system and the reed system was carried out, and the optimization results were applied in the devices’ design and implementation. The new structure improves the electromagnetic suction and reed force at the suction position. And the results from simulation show that the proposed method exceeds that of the back propagation neural network method in achieving the optimization goal and the new structure can effectively improve the performance of meter built-in load switch by avoiding the recoil phenomenon.     

基于循环平稳特性的时频分析法欠定盲源分离

基于二次时频分布的算法是解决欠定盲源分离问题的一种有效方法。不同于传统算法,针对循环平稳信号,借助分段平均的周期图法求解谱相关密度函数,并利用其实现Wigner-Ville时频分布的重构。计算信号时频分布矩阵并找出自源时频点,利用自源时频点对应的时频分布矩阵构建新的3阶张量模型。利用平行因子分解,直接实现源信号的分离。该算法不需要假设任意时频点的源数目,不大于混合信号数目。仿真实验结果表明,所提出的方法可以有效地抑制噪声,并且只需要一步即可实现源信号的恢复,避免“两步法”造成的误差叠加,提高了盲源分离的效率和性能。     

Reconstruction of strain distribution in fiber Bragg grat-ings with differential evolution algorithm

Differential evolution algorithm is used to solve the inverse problem of strain distribution in tibet Bragg grating (FBG).Linear and nonlinear strain profiles are reconstructed based on the reflection spectra. An approximate solution could beobtained within only 50 rounds of evolutions. Numerical examples show good agreements between target strain profilesand reconstructed ones. Online performance analysis illuminates the efficiency and practicality of differential evolutionalgorithm in solving the inverse problem of FBG.     

Purging compounds reduce machine downtime and increase productivity for compounders

Effective cleaning of plastic processing equipment is vital for keeping the machinery running productively and free from trouble. Josephine Komarmi of Dyna-Purge® Div. of Shuman Plastics Inc. discusses the issues involved when comparing manual cleaning or in-house remedies to the use of commercial purging compounds for compounders.