Progress in lattice Boltzmann methods for magnetohydrodynamic flows relevant to fusion applications

In this paper, an approach to simulating magnetohydrodynamic (MHD) flows based on the lattice Boltzmann method (LBM) is presented. The dynamics of the flow are simulated using a so-called multiple relaxation time (MRT) lattice Boltzmann equation (LBE), in which a source term is included for the Lorentz force. The evolution of the magnetic induction is represented by introducing a vector distribution function and then solving an appropriate lattice kinetic equation for this function. The solution of both distribution functions are obtained through a simple, explicit, and computationally efficient stream-and-collide procedure. The use of the MRT collision term enhances the numerical stability over that of a single relaxation time approach. To apply the methodology to solving practical problems, a new extrapolation-based method for imposing magnetic boundary conditions is introduced and a technique for simulating steady-state flows with low magnetic Prandtl number is developed. In order to resolve thin layers near the walls arising in the presence of high magnetic fields, a non-uniform gridding strategy is introduced through an interpolated-streaming step applied to both distribution functions. These advances are particularly important for applications in fusion engineering where liquid metal flows with low magnetic Prandtl numbers and high Hartmann numbers are introduced. A number of MHD benchmark problems, under various physical and geometrical conditions are presented, including 3-D MHD lid driven cavity flow, high Hartmann number flows and turbulent MHD flows, with good agreement with prior data. Due to the local nature of the method, the LBM also demonstrated excellent performance on parallel machines, with almost linear scaling up to 128 processors for a MHD flow problem.     

Multiprocessor scheduling with communication delays

This paper adresses certain types of scheduling problems that arise when a parallel computation is to be executed on a multiprocessor. We define a model that allows for communication delays between precedence-related tasks, and propose a classification of various submodels. We also review complexity results and optimization and approximation algorithms that have been presented in the literature.     

Mesomixing scale controlling and its effect on micromixing performance

To direct highly efficient microdevice design, the mixing performance of different mixing methods was investigated. Three different microstructured mixers representing three kinds of mixing methods were utilized in this work. The mesomixing scale was adjusted through different ways in these mixers and the micromixing performance was characterized by a parallel competing reaction. A dimensionless parameter of mesomixing scale with different forms in these mixers was defined by considering contacting surface and mixing volume, and its relation with the segregation index which characterized the micromixing performance was investigated. All results indicate the enhancement of the micromixing performance with the decrease of the mesomixing scale. The mixing potential of different mixing method was discussed and it shows that the droplet cross-flow mixing method has the highest mixing potential. A linear relationship between the dimensionless parameter of mesomixing scale and the segregation index has been obtained. The results could provide much better understanding of how mesomixing scales affect the micromixing performance, which are very helpful for designing new micromixing devices and optimizing the geometric structures and operation conditions.     

Hard combinators

Hard Interaction systems can be presented as graph relabeling with a handshake mechanism that provide local synchronization. We present a particular one with only four symbols and seven rules that can be used to simulate all the other hard interaction systems.     

Evaluating the mechanical injury problem in the wood-bamboo furniture manufacturing industry

This paper introduces the method and procedure of evaluating mechanical injury problems in the wood-bamboo furniture manufacturing industry in Taiwan. Epidemiological surveillance and scenario analyses were performed on 1695 workplace-related accidents to identify the general problem areas. Subsequently, on site investigations were conducted in 15 factories to determine the specific contributing factors of the two major hazard patterns in that industry: caught in/between as well as cut and laceration accidents. For the primary cause of the accidents, approximately 60% of the incidents were due to unsafe conditions, and approximately 40% of the incidents were attributed to unsafe acts. Two intervention measures are proposed to reduce the potential hazards resulting from unsafe conditions and unsafe acts.     

Elliptic curve cryptography on embedded multicore systems

The increasing use of network-connected embedded devices and online transactions creates a growing demand of network security for embedded systems. The security requirements, such as authentication, confidentiality and integrity, always make computationally intensive processes and can easily become the bottleneck of the related applications. In this paper we implement Elliptic Curve Cryptography (ECC) (Miller in Lecture Notes in Computer Science, vol. 218, pp. 417–426, 1985; Koblitz in Math. Comput. 48:203–209, 1987) on an embedded multicore system, and explore the task scheduling methods in different levels. First, we propose an instruction scheduling method that utilizes all the cores to perform one modular operation in parallel. Second, we perform multiple modular operations with multiple cores in parallel. The performance of those two implementations is compared and a scheduling method combining these two types of parallelism is proposed. We discuss the details of our proposed method by using an FPGA implementation of ECC over a prime field.     

Estimation of thermal and flow fields due to natural convection using support vector machines (SVM) in a porous cavity with discrete heat sources

Support vector machines (SVM), a soft programming technique, has been used to estimate the temperature distribution and flow fields in a square porous enclosure heated discretely by three isothermal heaters from the left vertical wall. Right vertical wall of the cavity was isothermal but it has colder temperature than the heaters while remaining walls were adiabatic. A database was prepared by solving the governing equations which were written using Darcy flow model. Using finite difference method to discretize the equation, a computational fluid dynamics (CFD) code was written. A correlation was developed between Nusselt and Rayleigh numbers. Using obtained database, further values of temperature and velocities were estimated by SVM technique at different Rayleigh numbers and locations of heater. It was observed that SVM was a useful technique on estimation of streamlines and isotherms. Thus, SVM reduces the computational time and helps to solve some cases when CFD fails to solve due to numerical instability.     

Predicting metamorphic relations for testing scientific software: a machine learning approach using graph kernels

Comprehensive, automated software testing requires an oracle to check whether the output produced by a test case matches the expected behaviour of the programme. But the challenges in creating suitable oracles limit the ability to perform automated testing in some programmes, and especially in scientific software. Metamorphic testing is a method for automating the testing process for programmes without test oracles. This technique operates by checking whether the programme behaves according to properties called metamorphic relations. A metamorphic relation describes the change in output when the input is changed in a prescribed way. Unfortunately, finding the metamorphic relations satisfied by a programme or function remains a labour‐intensive task, which is generally performed by a domain expert or a programmer. In this work, we propose a machine learning approach for predicting metamorphic relations that uses a graph‐based representation of a programme to represent control flow and data dependency information. In earlier work, we found that simple features derived from such graphs provide good performance. An analysis of the features used in this earlier work led us to explore the effectiveness of several representations of those graphs using the machine learning framework of graph kernels, which provide various ways of measuring similarity between graphs. Our results show that a graph kernel that evaluates the contribution of all paths in the graph has the best accuracy and that control flow information is more useful than data dependency information. The data used in this study are available for download at http://www.cs.colostate.edu/saxs/MRpred/functions.tar.gz to help researchers in further development of metamorphic relation prediction methods. Copyright © 2015 John Wiley & Sons, Ltd.     

8-SPU并联机器人工作空间分析及参数优化

并联机器人工作空间是评价并联机构工作能力的重要指标,而结构参数是影响工作空间的基本因素。通过圆弧相交法求定姿态的工作空间,得到其截面形状,利用梯形积分法计算出工作空间体积。以工作空间体积为目标函数,通过遗传算法对机构的结构参数进行优化,获得了性能优良的结构参数。利用MATLAB软件编程得到了该并联机构的工作空间图,可以更直观观察工作空间的变化,为该并联机构性能的进一步研究和推广应用奠定了基础。     

多机电作动器同步控制器设计与实现

机电作动系统是飞机电传操纵系统的执行机构,是飞行控制系统中的重要组成部分,是多电-全电飞机的发展方向。机电作动系统在驱动飞行器舵面时,需要多个(一般3个以上)机电作动器采用力综合方式工作,这样会产生不同步问题,鉴于此,提出一种改进型同步控制方案,并通过同步控制器实现同步控制策略。该控制策略综合主从和等同控制的各自优势,并通过优化比较器提高同步控制性能。控制器的实现则基于可编程逻辑单元,通过并行总线进行数据交互,将并行总线控制器与同步控制有效结合。实际应用表明:目前控制器工作良好,提高了现有同步控制器的实时性和可靠性。